J. vet. Pharmacol. Therap. 35 (Suppl. 2), 45–51. doi: 10.1111/j.1365-2885.2012.01409.x
`
`Naloxone reversal of an overdose of a novel, long-acting transdermal
`fentanyl solution in laboratory Beagles
`
`K. J. FREISE*
`
`G. C. NEWBOUND*
` 
`
`C. TUDAN
`
`&
`
`T. P. CLARK*
`
`*Nexcyon Pharmaceuticals, Inc., Madison,
` 
`WI, USA;
`BioAccurate Enterprises, Inc.
`Vancouver, BC, Canada
`
`Freise, K. J., Newbound, G. C., Tudan, C., Clark, T. P. Naloxone reversal of an
`overdose of a novel, long-acting transdermal fentanyl solution in laboratory
`Beagles. J. vet. Pharmacol. Therap. 35 (Suppl. 2), 45–51.
`
`Opioid overdose in dogs is manifested by clinical signs such as excessive
`sedation, bradycardia, and hypothermia. The ability of two different intramus-
`cular (i.m.) naloxone reversal regimens to reverse the opioid-induced effects of a
`fivefold overdose of long-acting transdermal fentanyl solution was evaluated in
`dogs. Twenty-four healthy Beagles were administered a single 13 mg ⁄ kg dose
`(fivefold overdose) of transdermal fentanyl solution and randomized to two
`naloxone reversal regimen treatment groups, hourly administration for 8 h of
`40 (n = 8) or 160 lg ⁄ kg i.m. (n = 16). All dogs were sedated and had reduced
`body temperatures and heart rates (HRs) prior to naloxone administration.
`Both dosage regimens significantly reduced sedation (P < 0.001), and the
`160 lg ⁄ kg naloxone regimen resulted in a nearly threefold lower odds of
`the 40 lg ⁄ kg i.m. naloxone regimen (P < 0.05).
`sedation than that of
`Additionally, naloxone significantly increased the mean body temperatures
`and HR (P < 0.001), although the 160 lg ⁄ kg regimen increased body
`temperature and HR more (P < 0.05). However, the narcotic side effects of
`fentanyl returned within 1–3 h following termination of the naloxone dosage
`regimens. The opioid-induced effects of an overdose of transdermal fentanyl
`solution can be safely and effectively reversed by either 40 or 160 lg ⁄ kg i.m.
`naloxone administered at hourly intervals.
`
`(Paper received 30 January 2012; accepted for publication 16 April 2012)
`
`Dr Terrence Clark, Nexcyon Pharmaceuticals, Inc., 644 W. Washington Ave.,
`Madison, WI 53703, USA. E-mail: clarktp@nexcyon. com
`
`INTRODUCTION
`
`The general recommendation for the selection and use of
`postoperative analgesics depends on the anticipated magnitude
`and duration of pain which in turn is influenced by the site,
`nature, and extent of surgery. In addition, both the character-
`istics of the analgesic and patient factors must be considered in
`the selection and continued use of an analgesic. The character-
`istics of an ideal analgesic have been considered and may
`include,
`in part, that the agent is a full agonist providing
`maximal analgesia for a wide range of pain states; has a rapid
`onset of action and a long duration of action; has linear kinetics;
`produces minimal adverse effects; is not vulnerable to important
`drug–drug interactions;
`is not significantly bound to plasma
`proteins; has no active metabolites; and it is reversible (Smith,
`2008; Moore, 2009). Many strategies have been pursued to
`identify technologies that meet these characteristics. At the drug
`discovery level, attempts have been made to identify an ideal
`analgesic by engineering the mu-opioid receptor (Tao et al.,
`
`Ó 2012 Blackwell Publishing Ltd
`
`2010). Others have concentrated on ways to alter existing
`analgesics or to combine existing analgesic compounds with
`drugs that may improve effectiveness while minimizing adverse
`effects (Smith, 2008).
`ideal analgesics and are
`Opioids have some features of
`generally regarded as an important part of multimodal
`postoperative analgesia, especially for moderate-to-severe pain.
`In veterinary medicine, there are limited approved products and
`extended extra-label use of orally administered opioids in dogs
`beyond the immediate postoperative period is prevented by
`inherent
`limitations including poor oral bioavailability and
`rapid clearance (Pascoe, 2000). As a result, extra-label opioid
`use is primarily limited to preoperative epidural or intrathecal
`injections, single or repeat parenteral
`injections or constant
`rate intravenous infusions delivered during anesthesia. To
`overcome these limitations and prolong the therapeutic dura-
`tion of action, other variations
`in opioid pharmaceutical
`delivery have been advanced for human or veterinary use that
`include extended release oral
`tablets
`(Holt
`et al., 2007),
`
`45
`
`Nalox1030
`Nalox-1 Pharmaceuticals, LLC
`Page 1 of 7
`
`

`

`46 K. J. Freise et al.
`
`transdermal patches (Hofmeister & Egger, 2004), and liposome-
`encapsulated injectable opioids (Smith et al., 2004). As a
`delivery method, the transdermal route has several potential
`strengths over oral and parenteral dose. These include nonin-
`vasive dosing, avoidance of the gastrointestinal tract, lack of
`first pass metabolism, steady, continuous drug delivery rather
`than a peak and trough phenomenon, potential reduction of
`side effects by elimination of peaks, possible reduction of lack of
`effectiveness owing to the elimination of troughs, and reduced
`dose frequency for convenience and increased compliance
`(Urquhart, 2000).
`Recently, a novel, long-acting transdermal fentanyl solution
`(RecuvyraÔ 50 mg ⁄ mL transdermal solution; Nexcyon Phar-
`maceuticals Ltd, London, UK) has been developed that
`potentially mitigates the disadvantages of oral, parenteral,
`and patch-delivered opioids and has several features of an ideal
`analgesic that include the following:
`fentanyl
`is a selective,
`l-opioid receptor agonist with a potency 100 times that of
`morphine (Stanley, 1992); it has a rapid onset of action and
`long duration of action, providing analgesic concentrations of
`fentanyl within 2–4 h of application for a duration of at
`least 4 days
`(Freise et al., 2012a,b);
`it has demonstrated
`dose-proportional plasma fentanyl concentrations following a
`single topical application (Freise et al., 2012a);
`there are
`minimum adverse effects at the selected dose with well-known
`opioid adverse events increasing in magnitude and frequency
`when administered up to five times the dose (Savides et al.,
`2012).
`As an ideal drug feature, reversibility has not been examined
`with transdermal
`fentanyl solution. Reversibility allows clini-
`cians to terminate the clinical effects of a drug when they are no
`longer deemed necessary to case management and permits
`intervention in the event of an overdose. Naloxone is an FDA
`approved opioid antagonist (NADA 035-825) that is considered
`the fentanyl reversal agent of choice in dogs because, as a pure
`opioid receptor competitive antagonist,
`it does not have the
`respiratory side effects of other opioid antagonists (Adams, 2001;
`Plumb, 2002). It has the highest affinity at the l-opioid receptor
`and successfully reverses the effects of fentanyl citrate injections
`in the dog (Paddleford & Short, 1973; Veng-Pedersen et al.,
`1995; Adams, 2001).
`The outcome of a fivefold overdose of transdermal fentanyl
`solution has been described and includes, in part, moderate-to-
`severe sedation, reduced rectal temperature, and reduced HR
`(Savides et al., 2012). It is likely that opioid-induced adverse
`events are reversible; however, given the long duration of action
`of transdermal fentanyl solution, it remains to be determined the
`duration of reversal
`from a single injection of naloxone.
`Therefore, the objective of this study was to determine an
`intramuscular (i.m.) naloxone reversal regimen to the opioid-
`induced effects from an overdose of
`transdermal
`fentanyl
`solution in dogs. To achieve the objective, two different i.m.
`naloxone doses were administered at hourly intervals and were
`evaluated for the reversal of peak sedation, reduced rectal
`temperature, and reduced HR in Beagle dogs following the
`
`administration of a single fivefold overdose (13 mg ⁄ kg) of
`transdermal fentanyl solution.
`
`MATERIALS AND METHODS
`
`Animals and experimental design
`
`Twenty-four healthy (based on physical examination) purpose-
`bred laboratory Beagles (12 males ⁄ 12 females), 5–6 months of
`age and ranging in bodyweight from 4.35 to 8.20 kg were
`selected. Dogs were individually housed, fed a commercial dry
`food formula and allowed ad libitum access to water. The animal
`facility temperature was maintained between 18 and 29 °C with
`30–70% relative humidity. The 24 selected dogs were random-
`ized to two different i.m. naloxone treatment groups, 40 lg ⁄ kg
`(n = 8) and 160 lg ⁄ kg (n = 16). An unbalanced study design
`was utilized because based on pilot experiments it was suspected
`that the recommended 40 lg ⁄ kg i.m. naloxone dose (Plumb,
`2002) would not provide sufficient reversal at the administered
`dose of transdermal fentanyl solution. All dogs were administered
`a single fivefold (13.0 mg ⁄ kg) overdose (use dose of 2.6 mg ⁄ kg)
`of transdermal fentanyl solution (RecuvyraÔ 50 mg ⁄ mL trans-
`dermal solution; Nexcyon Pharmaceuticals Ltd) to the ventral
`abdomen using a proprietary applicator tip as previously
`described (Freise et al., 2012a). Prior studies demonstrated that
`plasma fentanyl concentrations were as high or higher with
`ventral abdomen application compared to the labeled dorsal,
`interscalpular region application (Freise et al., 2012a). As this
`was an intentional overdose study, the worse case was examined
`with the ventral abdomen application. Sixteen hours after the
`application of transdermal fentanyl solution, when near maximal
`side effects were expected to be obtained (Savides et al., 2012),
`i.m. naloxone was administered into the dorsal lumbar muscles
`hourly for eight doses according to the treatment randomization.
`Sedation assessments were conducted by blinded assessors as
`none or sedated. There was no attempt to distinguish mild,
`moderate, or severe sedation because a previous study with a
`fivefold transdermal
`fentanyl solution overdose demonstrated
`that all dogs were moderately or severely sedated (Savides et al.,
`2012). In addition, the objective, continuous response variables
`of rectal body temperature and HR were collected 5 min before,
`10 min after, and 40 min after each naloxone dose. Additional
`sedation, rectal
`temperature, and HR measurements were
`collected at )1, 0, 14, 15, 24, 26, and 28 h following
`transdermal
`fentanyl solution administration. Venous blood
`samples for plasma fentanyl and naloxone concentrations were
`also collected from each dog at 0 (prior to transdermal fentanyl
`solution administration), 16 (prior to the 1st naloxone admin-
`istration), 20.083 (5 min following
`the 5th naloxone
`administration), and 24 (1 h following the last naloxone
`administration) hours post-transdermal fentanyl solution admin-
`istration directly into sodium heparin blood collection tubes.
`Plasma was harvested by centrifugation at 1500 g for 10 min at
`5 °C. Plasma samples were stored at )70 °C until analysis. All
`
`Ó 2012 Blackwell Publishing Ltd
`Nalox1030
`Nalox-1 Pharmaceuticals, LLC
`Page 2 of 7
`
`

`

`Naloxone reversal of transdermal fentanyl solution 47
`
`detection was conducted using positive ionization mode and
`the transitions 337.2 m ⁄ z fi 188.3 m ⁄ z
`monitoring of
`for
`fentanyl, 342.2 m ⁄ z fi 188.3 m ⁄ z for the IS fentanyl-d5, m ⁄ z
`328.2 fi m ⁄ z 310.2 for naloxone, and m ⁄ z 342.2 fi m ⁄ z 324.2
`for the IS naltrexone. Standard curves were determined using
`linear and quadratic regression for fentanyl and naloxone,
`respectively, with 1 ⁄ x2 weighting using Watson v7.0.0.01
`(Thermo Fisher Scientific), where x is the nominal sample
`concentration. Typical squared correlation coefficient (R2) values
`were 0.9972 and 0.9964 for fentanyl and naloxone, respec-
`tively. Concentration calculations were based on the peak area
`ratios of fentanyl to fentanyl-d5 and of naloxone to naltrexone
`for fentanyl and naloxone, respectively. The intra- and inter-
`assay precision (i.e., coefficient of variation) was £ 8.6%, and
`the accuracy (i.e., relative error) ranged from )4.2% to 6.0%
`for both analytes. The lower limit of quantification (LLOQ)
`was 0.1 and 1.0 ng ⁄ mL for plasma fentanyl and naloxone,
`respectively.
`
`Statistical methods
`
`The sedation assessments were analyzed using a generalized
`linear repeated measures mixed effects model. The logit trans-
`formation of the mean probability of sedation, li, for the ith
`subject was linearly related to time as follows:
`
`
`l
`1 l
`i
`
`ð
`logit l
`
`i
`
`¼
`
`8>>>>>><
`>>>>>>:
`
`Þ  log
`þ bi
`b
`B
`þ b
`b
`B
`þ b
`b
`B
`þ b
`b
`B
`þ b
`b
`B
`
`i
`
`þ bi
`þ b40
`þ b160
`þ bi
`
`N
`
`N
`
`þ bi
`þ bi
`
`F
`
`F
`
`F
`
`F
`
`if t  0
`if 0 < t  16
`if 16 < t  24 and Dose ¼ 40 lg=kg
`if 16 < t  24 and Dose ¼ 160 lg=kg
`if t > 24
`
`N
`
`3
`
`N
`
`where t is the nominal time of observation in hours, bB, bF, b40
`N ,
`N , are the baseline, fentanyl, 40 lg ⁄ kg i.m. naloxone,
`and b160
`and 160 lg ⁄ kg i.m. naloxone fixed effect terms, respectively.
`Additionally, bi is the subject-specific random effect term that is
`normally distributed with mean 0 and variance r2. The overall
`narcotic reversal effect of i.m. naloxone administered at hourly
`intervals on sedation was tested with the null hypothesis of
` b40
`þ 2
` b160
` 0 vs. the alternative hypothesis 1
` b40
`þ
` b160
`N
`3
`
`
`N < 0. The unequal weighting of b40N and b160N was used to
`account for the unbalanced design of the study (n = 8 and
`n = 16 for the 40 and 160 lg ⁄ kg naloxone doses, respectively).
`The narcotic reversal effectiveness of each naloxone dose was
`‡ 0, b160
`‡ 0 vs. the
`tested with the null hypotheses that b40
`N
`N
`
`N < 0, b160N < 0. To test
`alternative hypotheses that b40
`the
`additional effect of 160 lg ⁄ kg vs. a 40 lg ⁄ kg i.m. naloxone
`doses administered at hourly intervals, the null hypothesis of
`‡ b40
`b160
`N was tested against the alternative hypothesis that
`N
`
`N < b40N . As 100% of dogs in both group 1 and 2 were sedated
`b160
`during the time intervals 0 < t £ 16 and t > 24, a large
`correlation (near )1.00) existed between bF and b40
`
`N , b160N ,
`resulting in large standard errors of the parameters estimates.
`The large degree of correlation results in many combinations of
`
`13
`
`23
`
`Nalox1030
`Nalox-1 Pharmaceuticals, LLC
`Page 3 of 7
`
`procedures were approved by the local Institutional Animal Care
`and Use Committee.
`
`Plasma sample analysis
`
`Plasma fentanyl and naloxone concentrations were analyzed by
`liquid chromatography–tandem mass spectrometry. In brief, a
`Ò
`stock solution of fentanyl (Cerilliant
`, Round Rock, TX, USA)
`Ò
`) was diluted in
`and a stock solution of naloxone (Cerilliant
`Ò
`, Morristown,
`50:50 methanol (Honeywell Burdick & Jackson
`NJ, USA):water (Milli-Q; Millipore Corp., Billerica, MA, USA) to a
`25 and 250 lg ⁄ mL working solution, respectively. Control dog
`plasma (Bioreclamation Inc., Hicksville, NY, USA) was then
`serially diluted with the working solution to create standard
`curve samples ranging from 0.1 to 10 ng ⁄ mL of fentanyl and 1
`to 1000 ng ⁄ mL naloxone. Additionally, an internal standard
`Ò
`(IS) working solution of fentanyl-d5 (Cerilliant
`) and naltrexone
`Ò
`) at concentrations 200 and 2000 ng ⁄ mL, respec-
`(Cerilliant
`tively, was prepared in 50:50 methanol ⁄ water. A 100 lL each
`of sample, standard, quality control, or control blank was
`aliquoted directly into a 96-well block, and 20 lL of the IS
`working solution was added to all wells except for the control
`blanks, which instead had 20 lL of 50:50 methanol ⁄ water were
`added and vortexed. Four hundred microliter of 5% acetic acid
`(Mallinckrodt Baker, Phillipsburg, NJ, USA) in water was then
`added to each well, and the samples were vortexed again
`followed by centrifugation at 4 °C. Solid-phase extraction (SPE)
`Ò
`then proceeded using Bond Elut
`96 Certify, 50 mg sample
`extraction blocks (Varian Corp., Palo Alto, CA, USA), and a
`Tomtec Quadra-96 Model 320 (Tomtech, Hamden, CT, USA).
`Sample blocks were conditioned twice with 400 lL of methanol,
`followed by equilibration with two 400-lL volumes of Milli-Q
`water and equilibration with two 400-lL volumes of 5% acetic
`acid in water. Samples were then transferred to the SPE block
`and slowly aspirated. The SPE block was then washed twice with
`400 lL of 5% acetic acid in water followed by a two washes of
`400 lL of methanol. Samples were slowly eluted from the SPE
`block with two 300-lL volumes of 2% ammonium hydroxide
`(EMD Biosciences, Darmstadt, Germany) in acetonitrile (Honey-
`Ò
`well Burdick & Jackson
`) and evaporated to dryness before
`reconstitution with 200 lL of 1% formic acid (EMD Biosciences)
`in acetonitrile.
`Reconstituted samples were quantified using an API 3000
`triple quadrupole mass spectrometer (Applied BioSystems ⁄ MDS
`SCIEX, Foster City, CA, USA) with peak area integration
`conducted using Analyst Software v 1.4 (Applied BioSys-
`tems ⁄ MDS SCIEX) data acquisition system. HPLC separation
`was achieved using a Thermo Betasil Silica-100 column
`(50 · 3 mm, 5 lm) (Thermo Fisher Scientific, Waltham, MA,
`USA) with the flow rate set at 0.5 mL ⁄ min. Mobile phase A
`consisted of 1% formic acid in water and mobile phase B
`consisted of 1% formic acid in acetonitrile. The mobile phase
`gradient started at 90% mobile phase B from 0.0 to 1.0 min,
`switched from 90% to 70% mobile phase B from 1.0 to 1.5 min,
`and switched back from 70% to 90% mobile phase B at 2.5 min.
`The injection volume was 10 lL, and mass spectrometer
`
`Ó 2012 Blackwell Publishing Ltd
`
`

`

`48 K. J. Freise et al.
`
`parameter estimates giving almost identical fits to the data. To
`alleviate this issue, the bF parameter was fixed to the initially
`estimated value (i.e., the estimate when all five terms in the
`model were simultaneously estimated), and then, the statistical
`analysis was conducted to determine the effect of naloxone on
`sedation, conditional on the fixed value of bF. The bF parameter
`was chosen to be fixed as it is already known that fentanyl has
`sedative effects (Freise et al., 2012b; Adams, 2001; Plumb,
`2002) and because the interest of the study was the effect of
`naloxone, not the effect of fentanyl. A sensitivity analysis was
`subsequently conducted to determine the effect on the statistical
`conclusions of changing the value of bF to other reasonable
`values.
`The body temperature and HR were analyzed with a linear
`repeated measures mixed effects model with dose, nominal time,
`and dose by nominal time interaction terms as fixed effects and
`subject as a random effect in the model. The covariance structure
`in the repeated measures analysis was investigated using three
`structural assumptions, namely compound symmetry, first-order
`autoregressive, and heterogeneous first-order autoregressive.
`The assumption that gave the minimum value of the Akaike’s
`Information Criterion was selected for the final model (Akaike,
`1974). For both body temperature and HR, the first-order
`autoregressive model was selected. The overall narcotic reversal
`effect of
`i.m. naloxone administered at hourly intervals was
`tested with a null hypothesis of lN £ lF, vs. the alternative
`hypothesis lN > lF, where lN is the mean body temperature or
`HR during the naloxone treatment time period (16 < t £ 24)
`and lF is the mean body temperature or HR during the fentanyl
`only time period (t = 14, 15, 15.917, 26, 28). To test the
`additional effect of 160 lg ⁄ kg vs. a 40 lg ⁄ kg i.m. naloxone
`doses administered at hourly intervals, the null hypothesis of
`£ l40
`l160
`N was tested against the alternative hypothesis of
`N
`l160
`
`
`N > l40N , where l40N and l160
`are the mean body temperatures
`N
`or HRs during the naloxone treatment time period for the 40 and
`160 lg ⁄ kg i.m. naloxone dose groups, respectively.
`All statistical analyses and calculations were conducted in
`SAS (version 9.1.3 Service Pack 4; SAS Institute Inc., Cary, NC,
`USA). The sedation scores were analyzed using the NLMIXED
`procedure, and the body temperatures and HR were analyzed
`using the MIXED procedure. Statistically significant differences
`the a = 0.05 probability of a type I
`were determined at
`experiment-wise error. To control the experiment-wise error
`rate, the unadjusted P-values were corrected using the step-
`
`down Bonferonni method for multiple tests on each response
`variable (Holm, 1979). Specific hypotheses were tested using the
`ESTIMATE statement in SAS and unadjusted P-values con-
`structed using a Student’s t-test.
`
`RESULTS
`
`The plasma naloxone and fentanyl concentrations are displayed
`in Table 1. Plasma fentanyl concentrations were below the LLOQ
`prior to dosing in all dogs, and the mean fentanyl concentrations
`ranged from 4.60 to 6.53 ng ⁄ mL across both groups from 16
`through 24 h following the administration of a fivefold overdose
`(13 mg ⁄ kg) of
`transdermal
`fentanyl
`solution. The plasma
`naloxone concentrations were also below the LLOQ prior to
`i.m. naloxone dose administration in all dogs. At 5 min following
`the 5th naloxone dose administration (20.083 h), the plasma
`naloxone concentrations were 10.4 ± 0.238 (mean ± standard
`error) and 34.7 ± 1.76 ng ⁄ mL in the 40 and 160 lg ⁄ kg i.m.
`naloxone dose groups, respectively. At 24 h, the mean plasma
`naloxone concentrations had dropped substantially from the
`previous peaks in both groups, consistent with its known short
`duration of action and rapid clearance (Veng-Pedersen et al.,
`1995; Adams, 2001; Plumb, 2002). No seizures or other adverse
`affects of naloxone administration were observed in any dogs.
`The observed proportions of dogs sedated vs.
`time are
`displayed in Fig. 1 for both the 40 and 160 lg ⁄ kg naloxone
`dose groups. As can be observed, the baseline, pretransdermal
`fentanyl solution administration proportion of sedated dogs is
`near 0.0. For nonapparent reasons, six dogs were scored as
`sedated at the time of transdermal fentanyl solution application
`resulting in the proportion of sedated dogs being 0.4 and 0.2 in
`the 40 and 160 lg ⁄ kg groups, respectively, at time 0. Following
`a fivefold overdose of transdermal
`fentanyl solution, all dogs
`were sedated prior to naloxone administration (i.e., at 14, 15,
`15.917 h). The administration of either 40 or 160 lg ⁄ kg i.m.
`naloxone at hourly intervals reduced the proportion of sedated
`dogs. The mean proportion of sedated dogs from 16 through
`24 h for the 40 and 160 lg ⁄ kg dose groups was 0.698 and
`0.438, respectively. Additionally, all dogs were determined to be
`sedated at least once from 16 through 24 h in both groups. The
`mean proportion of sedated dogs returned to 1.0 following
`cessation of the hourly i.m. naloxone administrations for both
`groups by 26 h.
`
`Table 1. Plasma fentanyl and naloxone concentrations by treatment group
`
`Plasma fentanyl conc. (ng ⁄ mL)
`40 lg ⁄ kg i.m. naloxone
`160 lg ⁄ kg i.m. naloxone
`
`Plasma naloxone conc. (ng ⁄ mL)
`40 lg ⁄ kg i.m. naloxone
`160 lg ⁄ kg i.m. naloxone
`
`Time (h)
`
`Mean
`
`Standard error
`
`Mean
`
`Standard error
`
`Mean
`
`Standard error
`
`Mean
`
`Standard error
`
`0
`16
`20.083
`24
`
`<LLOQ
`4.60
`5.92
`5.42
`
`–
`0.537
`0.873
`0.919
`
`<LLOQ
`5.46
`5.80
`6.53
`
`–
`0.295
`0.358
`0.568
`
`<LLOQ
`<LLOQ
`10.4
`2.78
`
`–
`–
`0.238
`0.201
`
`<LLOQ
`<LLOQ
`34.7
`12.8
`
`–
`–
`1.76
`0.848
`
`<LLOQ, Less than lower limit of quantification for all subjects.
`
`Ó 2012 Blackwell Publishing Ltd
`Nalox1030
`Nalox-1 Pharmaceuticals, LLC
`Page 4 of 7
`
`

`

`Naloxone reversal of transdermal fentanyl solution 49
`
`40 (cid:1106)g/kg
`
`160 (cid:1106)g/kg
`
`38.5
`
`37.5
`
`36.5
`
`35.5
`
`34.5
`
`Temperature (°C)
`
`16
`
`20
`
`24
`
`28
`
`–4
`
`0
`
`4
`
`8
`
`16
`
`20
`
`24
`
`28
`
`12
`Time (h)
`
`40 (cid:1106)g/kg
`
`160 (cid:1106)g/kg
`
`1
`
`0.8
`
`0.6
`
`0.4
`
`0.2
`
`Proportion sedated
`
`0
`
`–4
`
`0
`
`4
`
`8
`
`12
`Time (h)
`
`Fig. 1. Observed proportion of dogs sedated vs. time by i.m. naloxone
`treatment group.
`
`Fig. 2. Mean rectal body temperature vs. time by i.m. naloxone
`treatment group. Bars represent the standard error.
`
`40 (cid:1106)g/kg
`
`160 (cid:1106)g/kg
`
`140
`
`120
`
`100
`
`80
`
`60
`
`40
`
`20
`
`0
`
`Heart rate (bpm)
`
`–4
`
`0
`
`4
`
`8
`
`12
`Time (h)
`
`16
`
`20
`
`24
`
`28
`
`Fig. 3. Mean heart rate vs. time by i.m. naloxone treatment group. Bars
`represent the standard error.
`
`101 ± 3.31 bpm prior to transdermal fentanyl solution admin-
`istration to 64.2 ± 3.04 bpm following transdermal
`fentanyl
`solution administration (i.e., at time 14, 15, and 15.917 h).
`During i.m. naloxone reversal (i.e., from 16 through 24 h), the
`HR across both groups returned to the prefentanyl administra-
`tion HR with a value of 101 ± 2.41 bpm and then dropped
`again to an overall mean of 83.1 ± 3.31 bpm following
`termination of naloxone administration. The mean HR during
`time period (lN) was 28.9 ± 1.78 bpm
`naloxone treatment
`higher than the mean during the fentanyl only time period (lF)
`(P < 0.001). Finally, during the naloxone treatment time period,
`the HR was 9.97 ± 5.11 bpm higher in the 160 lg ⁄ kg i.m.
`naloxone dose group (104 ± 2.95 bpm) than in the 40 lg ⁄ kg
`i.m. naloxone dose group (94.4 ± 4.17 bpm, P = 0.0258),
`further
`indicating greater narcotic reversal effect of
`the
`160 lg ⁄ kg i.m. naloxone dosage.
`
`DISCUSSION
`
`This study demonstrates that the opioid-induced effects of up to a
`fivefold overdose of transdermal fentanyl solution in dogs can be
`successfully reversed through administrations of either 40 or
`160 lg ⁄ kg i.m. naloxone. Naloxone is an approved opioid
`antagonist for use in the dog where the recommended initial dose
`
`Nalox1030
`Nalox-1 Pharmaceuticals, LLC
`Page 5 of 7
`
`The overall effect of naloxone on reversal of the sedative effects
`of transdermal
`fentanyl solution was statistically significant
`(P < 0.001), as was the individual effect of
`the 40 and
`160 lg ⁄ kg i.m. naloxone reversal regimens (P < 0.001 for both
`regimens). The analysis also indicated that there was significant
`subject-to-subject variability in the sedation response (i.e., the
`probability that r2 > 0 was < 0.05). Furthermore, the reversal
`affect of the 160 lg ⁄ kg i.m. naloxone dose was significantly
`the 40 lg ⁄ kg i.m. naloxone dose
`greater
`than that
`for
`(P = 0.0132). The odds of a subject being sedated with a
`160 lg ⁄ kg i.m. naloxone dose was 0.353 (95% confidence
`interval [0.0327–0.674]) fold that of a 40 lg ⁄ kg i.m. naloxone
`dose. Owing to the high degree of correlation (near )1.00)
`
`
`between bF and b40N , b160N , the value of bF was fixed to the initial
`estimate of 11.5. Varying the fixed value of bF from 1 to 30 had
`no affect on the hypothesis test results (accurate numerical
`integration of the likelihood could not be achieved for values of
`bF > 30),
`indicating that the results are robust to different
`reasonable fixed values of bF.
`The mean rectal body temperatures vs. time by i.m. naloxone
`dose group are displayed in Fig. 2. The rectal body tempera-
`tures across both groups dropped from 38.4 ± 0.0976 °C prior
`to transdermal
`fentanyl solution administration to 35.1 ±
`0.0884 °C following transdermal
`fentanyl solution treatment
`(i.e., at time 14, 15, and 15.917 h). During i.m. naloxone
`reversal (i.e.,
`from 16 through 24 h), the body temperature
`across both groups was 37.7 ± 0.0578 °C. By 26 and 28 h
`the body temperatures returned to near prenaloxone adminis-
`tration values with an overall mean of 35.9 ± 0.0976 °C. The
`mean body temperature during naloxone treatment
`time
`period (lN) was 2.19 ± 0.0638 °C higher than the mean
`during the fentanyl only time period (lF) (P < 0.001). Addi-
`tionally, during the naloxone treatment time period, the body
`temperature was 0.412 ± 0.123 °C higher in the 160 lg ⁄ kg
`i.m. naloxone dose group (37.8 ± 0.0708 °C)
`than in the
`40 lg ⁄ kg
`(37.4 ± 0.100 °C,
`i.m. naloxone
`dose
`group
`P < 0.001),
`indicating greater narcotic reversal effect of the
`higher i.m. naloxone dose.
`The mean HRs vs. time by i.m. naloxone dose group are
`displayed in Fig. 3. The HRs across both groups dropped from
`
`Ó 2012 Blackwell Publishing Ltd
`
`

`

`50 K. J. Freise et al.
`
`is 40 lg ⁄ kg administered by i.m., i.v., or subcutaneous injection
`followed by repeated doses, as needed, with at least 2–3 min
`between doses. It has been previously demonstrated to have rapid
`onset of action within 1–2 min of injection and a 45-min–3-h
`duration of action following a 30 lgÆkg ⁄ min fentanyl constant
`rate infusion (CRI) (Veng-Pedersen et al., 1995; Plumb, 2002).
`Likewise in the current study, renarcotization of dogs to at or near
`prereversal
`levels following a fivefold overdose of
`fentanyl
`occurred between 1 and 3 h after termination of the hourly
`naloxone dosage regimens. Naloxone is generally considered to be
`a safe agent with a wide therapeutic window; however, extremely
`high doses may cause seizures through c-aminobutyric acid
`receptor antagonism (Adams, 2001; Plumb, 2002). In the present
`study, i.m. naloxone administration was safe and the reversal of
`the sedative, hypothermic, and bradycardic effects of transdermal
`fentanyl solution occurred within minutes. Although both
`naloxone regimens were effective, the hourly 160 lg ⁄ kg i.m.
`naloxone reversal regimen was more effective at reversing the
`opioid-induced effects of an overdose.
`Sedation was recorded in six of 24 dogs in both treatment
`groups immediately prior to transdermal
`fentanyl solution
`administration, even though plasma fentanyl concentrations
`were not detectable. Therefore, the pretransdermal
`fentanyl
`solution administration sedation observations were most likely
`due to the subjective nature of sedation assessments. For
`example, the sedation scale that was utilized did not distinguish
`between mild, moderate, and severe sedation and a possibility
`was that blinded assessors could have scored relaxed dogs as
`sedated. Because of the subjective nature of sedation assessments
`and the numerical difficulties that required one of the model
`parameters to be fixed,
`the objective continuous response
`variables of body temperature and HR were also analyzed. The
`analysis of body temperature and HR response data confirm the
`conclusions from the sedation assessments, that both the 40 and
`160 lg ⁄ kg i.m. naloxone regimens were effective at reversing
`the opioid-induced effects of transdermal fentanyl solution and
`that the 160 lg ⁄ kg reversal regimen was more effective.
`Intramuscular naloxone rapidly reversed the effects of fentanyl
`overdose within minutes as clearly documented from the sedation,
`body temperature, and HR observations 10 min following the first
`naloxone administration. These results support the idea that rapid
`results can be achieved by i.m. injection without the need for direct
`i.v. access. The duration of a single naloxone injection is also
`illustrated by the peak and trough nature of the reversal between
`hourly injections. This phenomenon is likely reflective of the rapid
`absorption and elimination of naloxone which has a terminal
`elimination half-life of 1.19 h in the dog (Pace et al., 1979). The
`short duration of naloxone to reverse a fivefold overdose is
`illustrated in the 40 lg ⁄ kg group, whereby frequently all dogs
`were sedated immediately prior to the next naloxone administra-
`tion. Body temperatures nearly returned to the pretransdermal
`fentanyl solution administration temperatures, particularly in the
`160 lg ⁄ kg treatment group. Across both groups, the mean body
`temperature during the naloxone administration time period was
`only 0.7 °C below the pretransdermal fentanyl solution admin-
`istration mean body temperature illustrating near complete
`
`reversal. Likewise, naloxone administration returned the mean
`HR to approximately the pretransdermal fentanyl solution mean
`HR, indicating near complete reversal of the bradycardia.
`Fentanyl-induced respiratory depression and its reversal by
`naloxone were not assessed in this study as a previous dog safety
`study using up to a fivefold overdose of transdermal fentanyl
`solution demonstrated that respiratory rates were maximally
`decreased only 30% (Savides et al., 2012). Furthermore, other
`studies have demonstrated that plasma fentanyl concentrations
`as high as approximately 80 ng ⁄ mL reduce the respiratory rate
`by only approximately 11 breaths ⁄ min (50%) in spontaneously
`breathing dogs (Arndt et al., 1984). Respiratory rate, oxygen
`consumption, and blood gases (pCO2, pO2, and pH) do not change
`further as concentrations increase above 100 ng ⁄ mL. Another
`study of fentanyl in dogs using a transdermal patch confirmed by
`blood gas analysis that sustained steady-state plasma fentanyl
`concentrations of approximately 2 ng ⁄ mL do not cause hypo-
`ventilation (Welch et al., 2002). When taken together, the data
`indicate that respiratory depression is a safety aspect of limited
`concern following fentanyl administration to dogs.
`The shorter duration of action of naloxone relative to
`transdermal fentanyl solution may necessitate repeat injections
`until an overdose is
`satisfactorily treated. Application of
`transdermal fentanyl solution to the skin results in sequestration
`of fentanyl into the stratum corneum and sustained absorption
`characterized by flip-flop kinetics with a half-life of approxi-
`mately 70 h (Freise et al., 2012a,b). As a result of the prolonged
`absorption into systemic circulation over a period of days, in